Research Grants & Funding | Thomas Blein

Drought and Heat Tolerance in Plants through AI-Assisted Long Noncoding RNAs Design – DHARP

Mon, 01 Sep 2025 00:00:00 +0000

Rice is one of the most important food crops globally, but its productivity is highly vulnerable to climate change, especially to increasing episodes of drought and extreme heat. These stresses often occur together and have a significant impact on yield and food security. While traditional breeding has contributed to improving stress tolerance, this process is slow and frequently constrained by trade-offs between growth and resilience. New strategies are needed to help crops respond quickly and efficiently to changing environmental conditions.

This project aims to explore the potential of long noncoding RNAs (lncRNAs), a relatively understudied class of RNA molecules that do not encode proteins but are increasingly recognized as powerful regulators of gene expression. Because lncRNAs can influence how plants respond to stress, often in subtle and finely tuned ways, they represent a promising new layer of control over key survival traits. However, predicting their function remains a major scientific challenge, as they lack the sequence conservation typical of protein-coding genes. Nonetheless, lncRNAs that share similar biological roles often contain short conserved sequence or structural motifs linked to their interactions with protein partners. However, the systematic detection and characterization of these motifs remains a significant computational challenge due to their subtlety, variability, and the complexity of RNA-protein interaction landscapes.

Artificial intelligence (AI) offers a promising solution for identifying patterns and features beyond primary sequence similarity. Therefore, we propose to use deep learning embedding to analyze the structure and patterns of lncRNAs in rice and Arabidopsis, identifying those most likely to be involved in gene regulation in the context of heat and drought response. By combining AI predictions with high throughput functional screening, we aim to pinpoint the most effective lncRNAs and understand how they work in the plant’s stress response networks. Our ultimate goal is to turn this knowledge into practical tools: using synthetic lncRNAs delivered through innovative RNA spray technology to help plants better tolerate high temperatures and water scarcity—without the need for genetic modification.

This project merges cutting-edge artificial intelligence methods with experimental high throughput plant biology to unlock new ways of boosting crop resilience. It aims to lay the foundation for a new generation of RNA-based, climate-smart agricultural solutions.

Inter-organ modulation of gene regulatory networks in plants, the example of the CUP-SHAPED COTYLEDON genes – Imogen

Sat, 01 Feb 2025 00:00:00 +0000

The CUP-SHAPED COTYLEDON (CUC) transcription factors are evolutionarily conserved key players directing growth and patterning during aerial plant organ development. However, the cellular and molecular mechanisms underpinning these developmental roles remain largely elusive. Our preliminary data point to additional, not yet characterized, multiple roles of the CUC genes during root development, which are apparently different from those they have in the aerial part. This raises the question whether the CUC genes have multiple roles or a unique, yet-to-identify fundamental function which may be modulated by the developmental context. The first objective of this project is to characterize the roles of the CUC genes during Arabidopsis root development. The second objective is to establish the CUC gene regulatory network (GRN) and determine its possible modulation by the developmental context. The third objective is to determine to what extent the apparently diverse developmental roles of the CUC genes may in fact hide a unique still-to-identify fundamental role. Together, this project will not only shed light on the roles of a group of central regulators of plant development, but will also provide insight into the malleability of GRNs between plant organs.

Non-coding RNAs regulatory networks controlling lateral root organogenesis across Brassicaceae – RESOLAT

Wed, 01 Jan 2025 00:00:00 +0000

Long non-coding RNAs (lncRNAs) are now recognized as regulators of gene expression, from chromatin conformation to protein translation, playing a key role in a variety of biological processes. However, their mode of action remains poorly understood, due to the lack of known characteristics that can determine their mechanism of action. Their integration within biological networks remains largely unknown, in particular lateral root development.

The multidisciplinary RESOLAT project aims to understand the impact of lncRNAs on the gene network regulating lateral root development in Brassicaceae. Using time-series RNA-seq data, we propose to reconstruct the gene regulatory network controlling lateral root development in eight Brassicaceae, integrating coding and non-coding genes (small and long). This approach will enable us to identify new interactions and regulatory pathways, with particular emphasis on lncRNAs.

To overcome the challenges associated with the functional characterization of lncRNAs, we propose an approach based on the conservation of their function. Based on lateral root development, a highly conserved process in Angiosperms, we will use conservation of position in the regulatory network as a proxy for conservation of function, i.e. regulation of orthologous genes in different species. By integrating data on genomic conservation (genomic synteny, sequence, secondary structure) and expression conservation during lateral root development, we will seek to better understand the conservation features leading to the functionality of long lncRNAs, including their ability to regulate orthologous genes within Brassicaceae.

Using deep learning approaches based on large language models (LLMs) that will combine sequence and network data, we plan to develop a predictive tool to determine whether a lncRNA acts as a regulator of gene expression level.

Finally, we will use functional approaches in several species to validate the function of conserved lncRNAs in the control of lateral root architecture. First, we will use a semi-high-throughput method derived from Perturb-seq to confirm the regulatory action of conserved lncRNAs in A. thaliana. Based on these results, we will further functionally characterize lncRNAs conserved in several Brassicaceae species to validate their role in controlling lateral root architecture. By modifying the expression dynamics of conserved regulators, we should be able to quantitatively modify root architecture and confirm the regulatory links established by gene network inference.

By comparing different species with a common lateral root organogenesis program, this project will reveal evolutionary conserved and specific mechanisms controlling lateral root development in Brassicaceae, including both wild and cultivated species. It will also enable us to identify functionally conserved lncRNAs and analyze the features underlying their conservation.